Thinking about MMOs

I've been promising some thoughts on MMO design in this blog for quite some time already, and now that I've finally started playing a little WoW again (with the impending release of Cataclysm and what not), I figure now is a pretty good time to share a couple of the ideas I've had over the years. The real impetus for this post came from playing through the pre-Cataclysm quests. I had just stepped away from them at the point that I started writing this (two days ago), and on the whole I was struck by how the narrative pressure that's building in the Warcraft universe was almost completely washed out by the basic format of questing which remains more or less unchanged. More after the jump.

I mean, the quest sequence should have been fairly affecting. I checked a bunch NPCs for their papers to screen for possible terrorists, I arrested a bunch of cultist infiltrators who were trying to blow up Orgrimmar, then infiltrated their camp to learn how to sabotage the attack they were planning, and finally disrupted said attack (or at least some aspect of it). In theory, this narrative sequence should have been compelling and exciting. The truth is that it made Fable 3 look like a paragon of immersive game design by comparison and that's really saying something given some of the deep immersion breaking flaws in that game (a subject for another post). Part of me knows that this is just a result of having played a lot of WoW. The quest mechanics were once totally immersion enhancing for me and I'm just terminally burnt out on them. Another part of me knows that MMOs are capable of something more, and that's what this post is about.

Idea one came to me while playing in the Alterac Valley (AV) battleground roughly four years ago. Those of you who know WoW PvP know that AV has been tweaked to death by Blizzard over the years, and for the most part they've just made it progressively more and more broken. Back in the day it used to run for 2 hours at a minimum (often exceeding 4), and it was pretty epic. It was busted as hell just like the PvP ranking system, but it was still a lot of fun and it really did give you the feeling of being in a massive fantasy battle. In those early days, the biggest problem with the experience for me was usually the lack of any real leadership on either side in most battles. Granted, the leadership issue was pervasive throughout the battlegrounds but the combined size of 40 person teams and the requirements for coordinated action in bringing closure to AV made it particularly hellish at times.

This gave me an idea for a way in which an MMO fantasy battle could have a functional leadership system, but it would involve a little bit of tweaking to the game (and differentiated UI) like the sort Yahtzee discussed in that Extra Punctuation post where he talked about a hybrid RTS model. Incidentally, this offers another instance in which Croshaw and I have been thinking in parallel lines about potential hybridization of game genres/mechanics. If this keeps happening I'm going to have to have some words with him (although I have to admit I have no idea what the content of that conversation would actually be).

At any rate, the core idea I had regarding BGs revolved around the idea that an individual would be designated as the field general for a battleground prior to the start of each battle. That player would have the ability to fight as usual, but would also be able to drop into a map based interface in which they would have both moderate control of friendly NPC troop movement, and the ability to send directives out to the rest of the friendly players. In this way, the general would have real power. As general, in flanking an enemy advance you could both move NPCs into position and provide a simultaneous directive to the human players on their maps. I'm thinking the map would flash across the player's screens with arrows indicating the desired troop movements. Other players on your side would be highly motivated to play along with the call you've made so as to gain the advantage of the NPCs that you as general have mobilized.

Needless to say, there would have to be some mechanism for not only determining who would start each battle as general, but also for changing up the leadership during battle if players were dissatisfied with it. If the idea of a player leader seems too unwieldy, another option would be to have the general actions taken by the UI in response to player initiated changes on the map. This would be a more problematic approach when it comes to tuning and balancing game play, but it could potentially still create motivation for players to operate more collaboratively in a battleground setting on a large scale. For instance, capturing specific objectives would trigger reactions from the AI general to initiate NPC troop movements and the accompanying commands to the players. As before, players could choose to ignore the movements of their NPC allies but moving with them and supporting them would greatly enhance opportunities for victory. In order for this sort of solution to not become stagnant rather quickly, there would need to be a wide variety of objectives players could achieve that would initiate different strategic responses from the NPC generals.

A related PvP idea I had also taps into the battleground experiences I've had in Alterac Valley over the years. Specifically, the process of collecting and turning in items in AV (e.g. scraps of armor, blood and flesh of your enemies, etc.) for auxiliary troops and other aid generally has always felt like a low reward activity and a bit of a waste of time (especially as the BGs have grown shorter). The player turns all of this stuff in, and then if enough players work in concert you trigger some one off release of auxiliary troops that seldom impacts the battle in any meaningful way (I'm looking at you Frostwolf riders). At some point it struck me that there is a relatively easy way to retune the collection quest/auxiliary unit interaction to make it more meaningful for the players and more relevant for the outcome of the battleground. Instead of releasing an independent unit that strikes out under AI control, turning in an adequate number of items could instead grant the player a small number of NPC soldiers under their control. At a minimal level, this could be four or five soldiers that the player controls through a pet interface with all of the units responding as one. A more complex approach could involve utilizing a Sacrifice style unit placement/management system so that players could make meaningful tactical decisions with the small group of soldiers under their command.

Moving on, my second major idea around MMO structure also came to me roughly four years ago and it's focused on the intersection of in-game economies, virtual geography, played narratives, and questing structures. Like my thoughts on battlegrounds, this thread is also occupied with ways of ramping up the investment the player has in playing with the game mechanics instead of at the edges of them. The core idea is not entirely unrelated to the sort of phasing that Blizzard incorporated into WoW during WoTLK (which was suspiciously similar to the approach used by Turbine with LotRO), and it's also not entirely divergent from the Cataclysm which is taking place today in Azeroth. That is to say, this is an idea about locations in world changing and players gaining access to activities or locations which they couldn't prior to attaining certain objectives. This is the point where the similarity ends however. The core of my concept does involve changes that happen to the various settlements in game making different content available to players. However, rather than these events taking place as relatively tightly scripted narrative paths that advance the player through a primary narrative arc, these state changes would be optional, non-durable, and ultimately inconsistent between servers.

Imagine if every major village, town, or city in an MMO had at minimum three states that it could occupy. These three states would be something like dilapidated, healthy, and thriving. On a new server each location would be set to a default state that ensured a feeling of consistency for a world with only low level heroes. As players interact with the NPCs in each town they have the capacity to transform them by enhancing trade or completing very specific quest chains or possibly repeatable quests. Equally if all of the players on a server neglect a functioning town for an extended period of time, it starts to fall apart. As each location scales up (or down) a different array of shops, quests, and other infrastructural features become available (or disappear) as players meet or ignore various needs expressed by NPCs. If the game has a primary arc, quest givers in various locations could remain persistent despite other changes to locations. On the other hand, designers could force players into higher states of collaboration by requiring certain locations to be advanced in order for essential quests to open up, kind of like how Blizzard did with the Gates of Ahn'Qiraj and Fury of the Sunwell, only based around meaningful player activity rather than quotas that feel totally arbitrary. Equally, in a game defined around PvP, cities could be damaged by siege, captured, or if they were non-essential even razed. In essence there are a huge range of state changes that could be effected in a given virtual geography based on the setting and the type of narrative space the designers are seeking to immerse the players in. The trade off is a lack of consistency between player experiences on various servers, but it seems pretty clear that not all players are seeking a cookie cutter experience in their MMOs, even if many seem satisfied with it.

The other key feature of this system involves the mechanics utilized to advance those various settlements. State changes to locations in and of itself would be an interesting mechanic, but it only gets the player part way there in terms of feeling like the actions they take with NPCs and through questing actually matter. While repeatable quests are all well and good, they tend to emphasize to the player that they are engaged in a non-linear narrative loop which for me just makes the grind feel, well . . . more grindy. Meanwhile the activity of farming and vending also contributes to the feeling that the player is going through the motions to obtain virtual currency, but that the virtual objects themselves don't really have any meaning in world. This is particularly true of gray items, but it is more broadly true of any item that the player sells to a vendor in WoW and any similar RPGs (MMO or otherwise) as they just disappear into the ether.

The first premise I would like to proceed from is that the only items vendors purchase are items that somehow make sense for them to purchase in their setting. This might mean that items drop in one location which are only vendable in another location far away. Players who don't care about that other location don't bother to pick up those drops (thus alleviating the constant backpack management that attends the usual use of a DikuMUD style drop system). Apart from the complication of locations, the core idea is that every item that drops is actually meaningful within the game. If there's a potion vendor in town who collects bird bones, it's because that vendor actually uses those bones to make a specific type of healing potion. Bring that vendor bones and in addition to getting a little virtual currency, you also up the number of potions in stock or possibly even get a non-standard potion. If you find a bunch of low quality weaponry, it should be vendable to the local blacksmith who will break it down for spare parts. For player convenience other vendors might be willing to buy those weapons, but they'll be paying a lower price because their plan will be to go and sell those materials to the blacksmith themselves (nefarious little NPCs that they are). Ultimately not every item should be immediately relevant for goods and services players seek, but ever item that can be vended should be narratively consistent in relation to the game world.

In so far as repeatable quests go, there's no reason (apart from time and development cost) that they should give players an identical experience every time. If one village needs caravan escorts in order to enhance their economy, then players should be escorting a different caravan transporting different goods and susceptible to different dangers depending on how many caravans have been escorted in total, how many times the player has done the quest, and other conditions like time and weather. Equally, if the player's task is to train new recruits for the town guard, the nature of those recruits should vary based on how many have been trained so far. If the town is some significant portion of the way towards completing the target for this condition which will enable an upgrade if achieved in combination with several other conditions, perhaps the recruiting effort has started to hit the bottom of the barrel. Players might engage in the same basic activity with the last round of recruits as with the first, but successfully completing the activity could be significantly harder or simply require more repetition (e.g. Throk says, "This recruit can't wield a spear to save her life, better take her through the motions again."). Changes like these wouldn't take the grind out of the game, but they would repurpose parts of the grind to have narrative meaning inside of the game world.

While that doesn't exhaust my well of ideas for MMOs, it hits on two of the big ones (and some supporting concepts) that I've been stewing over for years. I expect that with Cataclysm I'm going to find more of the same WoW, and that it will be fun for a time based on the company I'll be keeping and the opportunity to see some of the changes Blizzard has made to the geography of Azeroth. I don't really expect it to be a more engaging MMO than the one I've been playing since 2005, and unfortunately after playing on Sunday I don't really expect it to live up to the narrative potential that the events unfolding in Azeroth would seem to promise. At the end of the day though that's alright. After all, if Blizzard released a game with the kinds of interactions in it that I'd like to see in an RPG I'd probably be back to where I was in June of 2005 spending the wee hours of the morning grinding out quests and collecting virtual junk to sell to NPCs, and then straggling in to work on less than 4 hours of sleep. Additionally, it's usually the moments of frustration and disappointment with games that give me the most inspiration for the games that could be, which is of course the fuel for a post like this one.